Comparison between human and artificial neural network detection of Laplacian-derived electroencephalographic activity related to unilateral voluntary movements

A back-propagation artificial neural network (ANN) was tested to verify its capacity to select different classes of single trials (STs) based on the s patial information content of electroencephalographic activity related to v oluntary unilateral finger movements. The rationale was that ipsilateral an d contralateral primary sensorimotor cortex can be involved in a nonstation ary way in the control of unilateral voluntary movements. The movement-rela ted potentials were surface Laplacian-transformed (SL) to reduce head volum e conductor effects and to model the response of the primary sensorimotor c ortex. The ANN sampled the SL from four or two central channels overlying t he primary motor area of both sides in the period of 80 ms preceding the el ectromyographic response onset in the active muscle. The performance of the ANN was evaluated statistically by calculating the percentage value of agr eement between the STs classified by the ANN and those of two investigators (used as a reference). The results showed that both investigator and ANN w ere capable of selecting STs with the SL maximum in the central area contra lateral to the movement (contralateral STs, about 25%), STs with considerab le SL values also in the ipsilateral central area (bilateral STs, about 50% ), and STs with neither the contralateral nor bilateral pattern ("spatially incoherent" single trials; about 25%). The maximum agreement (64-84%) betw een the ANN and the investigator was obtained when the ANN used four spatia l inputs (P < 0.0000001). Importantly, the common means of all single trial s showed a weak or absent ipsilateral response. These results may suggest t hat a back-propagation ANN could select EEG single trials showing stationar y and nonstationary responses of the primary sensorimotor cortex, based on the same spatial criteria as the experimenter. (C) 2000 Academic Press.